Well, it depends what point you're making. The preponderance of Ivermectin data suggests non-random repeatable effect sizes at preventing covid and preventing serious outcomes. It is incredibly myopic to respond to that data by pointing out that the sample sizes are small, because the meta-analysis sample sizes are large, and that's also…
Well, it depends what point you're making. The preponderance of Ivermectin data suggests non-random repeatable effect sizes at preventing covid and preventing serious outcomes. It is incredibly myopic to respond to that data by pointing out that the sample sizes are small, because the meta-analysis sample sizes are large, and that's also not how statistics works. Far too many people who don't really understand stats (not implying you) don't understand when the sample size criticism is warranted.
If you don't want to take any one study at face value, that's fine, there are hundreds.
One large sample study is not as strong as a hundred independent studies of smaller sample size. See: repeatability crisis, P-hacking, and so on.
Well, it depends what point you're making. The preponderance of Ivermectin data suggests non-random repeatable effect sizes at preventing covid and preventing serious outcomes. It is incredibly myopic to respond to that data by pointing out that the sample sizes are small, because the meta-analysis sample sizes are large, and that's also not how statistics works. Far too many people who don't really understand stats (not implying you) don't understand when the sample size criticism is warranted.
If you don't want to take any one study at face value, that's fine, there are hundreds.
One large sample study is not as strong as a hundred independent studies of smaller sample size. See: repeatability crisis, P-hacking, and so on.